Data is useless if it doesn’t shed light. The more light it sheds on the most acute problems businesses face, the better.
Within this context, data synergy–data from multiple sources and disciplines that is more valuable than the sum of its parts–is often underappreciated. With data synergy, the light can be in many more places, and can illuminate dark, unexplored nooks and crannies, providing previously undiscovered insights. Those insights constitute intelligence.
The intelligence community, when it has succeeded, has done so with the help of data synergy. How the US handled the Cuban Missile Crisis was an example of that synergy, with signals intelligence and imagery intelligence both playing a role. Signals intelligence provided an initial heads up regarding heightened activity in and around Cuba. Then focused imagery intelligence supplied details concerning Soviet missiles under construction there.
When the intelligence community has failed, the reasons have been because of a lack of data synergy and disconnectedness. An investigation after the September 11, 2001 (9/11) suicide terrorist attacks revealed, for example, that intelligence and criminal investigation data that could have been combined for early warning was instead kept separate, which implied that the only means of getting info from one system to another involved human intervention.
Because data has become so much more valuable, some private sector businesses are starting to resemble intelligence agencies. Let’s explore some of the distinctive features of intelligence and its capacity for data synergy and thoughtful exploration, by contrast with the narrow focus of business intelligence.
The known knowns matrix…and the unknowns
This past February 12th, we celebrated the 20th anniversary of a famous observation made three months after the 9/11 attack:
Reports that say that something hasn’t happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don’t know we don’t know. And if one looks throughout the history of our country and other free countries, it is the latter category that tends to be the difficult ones.
–Donald Rumsfeld at a Pentagon briefing, February 12, 2002.
The “known knowns” matrix Rumsfeld described has been popular ever since. In 2005, science historian Michael Shermer even wrote a Scientific American column about the scientists who’ve been influenced by this matrix. If you don’t ask good questions, obtain facts to help you answer those questions and understand what you don’t know and why you can’t make good decisions.
Alon Kiriati, FreeCodeCamp, 2019
Just using the known knowns matrix, of course, won’t guarantee good decisions. Several US administrations, for example, have each made their share of poor decisions after 9/11.
Groupthink–which happens when a group jumps to a quick consensus without weighing alternatives or considering consequences–is a prime reason poor decision-making in public and private sectors can occur so frequently.
The key to good decision-making is a divergent research phase that enables companies to discover different questions to ask–the known unknowns. Then establishing facts with a convergent phase that answers better questions can surface concerns to avoid pitfalls.
What is intelligence, really?
I did intelligence collection as a US Navy aircrewman back when I began my career. It was all about gaining a clearer understanding of military operations–of our enemies–by collecting what we could that was relevant about them, analyzing the data collected, and reporting on that data, in as close to real-time as we could.
The intelligence community I experienced was very systematic. We managed data throughout its entire lifecycle and shared what we found out via the appropriate channels. Those closer to the heart of the intelligence effort studied many different sources for more strategic insights: not only communications but other signals, imagery, human intelligence, open-source (public media), and even “fisint” (the emissions of radioactive material). Those folks would use one or more sources to confirm the information of an additional source. Once that was done, we’d get information back that helped us understand what we were collecting more.
There were feedback loops, in other words. The better the information gathering and analysis, the less bias in the reporting.
Business intelligence is getting broader
Let’s admit it: as more and more of the most useful information gets converted into machine-readable and processable form, more business organizations are starting to look like intelligence agencies. At least the online businesses are.
To function in the 21st century and stay competitive, all businesses have to be aggressive about the data they collect and methodical in how they manage it. And to be able to share what they collect, that data has to be contextualized so that machines can read it and the larger community can reuse it. Then businesses share the resulting knowledge (information with more context and clarity built into it) with designated knowledge consumers.
But unless a divergent and then convergent information collecting phase occur in sequence together, the knowledge decision-makers tap into can lack insights that would be critical to a decision.
Meanwhile, as business gets more and more interdependent, demand continues to grow for data sharing. And as data and enterprise architecture evolve, knowledge environments as a whole can grow organically as a semantic graph does–encouraging more connections, and divergent as well as convergent observations and thinking.